1. Technical Field
The present invention relates to an image scanner printer that is able to correct red-eye in an image read by a scanner and then to print out the image.
2. Related Art
In recent years, stand-alone printers, which can be used without being connected to a computer, become widely used. Some printers of this type have a slot for connecting a memory card or memory stick for digital camera, and others have installed an image scanner.
Some stand-alone printers are able to correct red-eye when they print out an image (JPEG image, RGB image, or the like) stored in a memory card. Red-eye is a phenomenon in which, when a human picture is taken with a flashlight in a dark place, the pupils of that person in the picture are sensitized to appear red (or gold). When a scanner is connected to a computer, and when a human image read by the scanner includes red-eye, it is possible to correct the red-eye with red-eye correction software that is installed in the computer.
The red-eye correction requires detection of a red-eye area and color correction of the red-eye area. Therefore, in advance of printing, the positions of eyes are pinpointed by recognizing a human face in an image, and it is determined whether red-eyes are present. When the red-eyes are present, their positions (red-eye area) are stored. Then, when printing, the color in the red-eye area of print data is replaced by black color, or the like.
In the stand-alone printer, when the red-eye correction is executed on an image stored in a memory card, image data are called from the memory card in units of a predetermined number of lines or in units of predetermined image blocks to an image processing portion. In this case, even when the capacity of a memory installed in the stand-alone printer is small, and even when no large work area can be ensured in the memory, it is possible to execute the red-eye correction by calling the image data little by little from the memory card.
On the other hand, a technology is known, in which when a plurality of photographs, or the like, including human faces are collectively read by one scanning operation, and when the layouts of images are incorrect (that is, when an image is tilted or out of position to the upper, lower, left, or right side), the layouts are automatically corrected, and, when red-eye is present, red-eye correction is executed and then printing is performed.
When the red-eye correction is executed on the image read by the scanner that is installed in a stand-alone scanner printer, the read image is once stored in a memory in the form of an image of RGB, or the like, and then red-eye detection and red-eye correction will be executed as in the case describe above.
However, in this case, the stand-alone scanner printer needs to have a large capacity memory, so that when the capacity of an internal memory is small, it is impossible to install a red-eye correction function.
When no large capacity memory is installed in a stand-alone scanner printer, by performing scanning at least twice with a scanner, it is not impossible to execute red-eye correction on an image read by the scanner.
For example, it may be possible that, only the detection of the position of red-eye is executed in a first scan (pre-scan), and red-eye correction is executed by referring to information regarding the detected position of red-eye in a second scan while data, on which the red-eye correction has been executed, are transferred to a printing mechanism.
Actually, when the second scan is executed, however, the red-eye area that has been detected in the first scan may vary from the red-eye area of an image that is read by the second scan because of a mechanical tolerance, or the like, of the scanner. As a result, there produces a problem that the red-eye correction is not properly executed and, therefore, a portion, which is not a red-eye portion, is replaced by black color, or the like. For the above reasons, in the existing stand-alone scanner printer, red-eye correction is not executed on an image that is read by the scanner.
Moreover, in a printing method, by which images are printed by automatically correcting the layouts of the images when the layouts are incorrect while executing red-eye correction for a picture including red-eye, because a normal printer has a small memory capacity, even when layout correction may be executed, there is a possibility that red-eye correction cannot be performed due to the memory capacity when red-eye is included in a picture (read image). On the other hand, even with a printer that is able to execute red-eye correction, when the printer tries to correct the incorrect layout of an image, which occurs when the scanner reads the image, there is a possibility that a red-eye area becomes out of position.